资源环境科技发展态势分析平台

COVID-19 exposed how underprepared the United States was for a pandemic and raised questions about how prepared it will be for a future pandemic. With the Delta variant surging and memories of last winter's wave still fresh, there appears to be political will to spend money on public health—and no shortage of suggestions on how to spend it.

Every vendor pushes their product. Every academic, association, and author calls for investment in their respective specialty. Each proposal may sound good in isolation, but how can the country take a holistic view of all the options? How should investments be prioritized? Simply spending more is not enough. Smarter spending is needed.

To achieve this, decisions should be based on solid data and analysis. Public health and policymakers could look broadly for solutions, beyond their usual modes of thinking. One possible approach, borrowed from the engineering world, could be failure mode analysis, which systematically examines potential ways a system can fail and prioritizes the fixes.

Specifically, failure mode analysis outlines the key functions needed in a pandemic response. It then identifies which of those steps can fail—or in this case, did fail. Next, it estimates the consequences of each specific failure. And finally, it identifies the potential for fixing the problems.

Functions involved in a pandemic response include contact tracing, testing, isolating infected persons to prevent spread, treating patients, developing a vaccine, and vaccinating people. The first function in the list, contact tracing, could be a good place to start, noting that health departments struggled to recruit and train enough contact tracers which made it difficult to identify everyone each infected person came in contact with. This is an area that could be fixed.

But given this, what would be the best use of public health dollars? A great deal of effort has led to the development of numerous mathematical models of the disease and public health strategies—models that could shed some light on this question. For example, models might show that for slower-moving diseases, contact tracing coupled with isolating exposed persons could effectively slow disease spread, but for faster-moving diseases, the number of tracers needed to outrace the spread would be too high to be practical. This suggests a limit on how much to invest in contact tracing, that additional resources might be better spent elsewhere, or at a minimum that flexibility should be built into the system.

So why don't we do such analysis regularly? First, it takes dedicated time and effort—resources in limited supply for agencies currently focused on responding directly to the pandemic. Doing this analysis right would require a review of the existing evidence, both scientific and operational, and engaging stakeholders with diverse expertise and experiences. Second, it would require a high-level view, agnostic of traditional silos. Pandemic response has many layers, drawing not only on biological and health sciences, but also engineering (e.g., logistics), social science (e.g., human behavior), and others. Modeling these diverse and dynamic factors would require an interdisciplinary approach quite foreign to many dedicated teams.

If done well, such an analysis could provide an analytic basis for prioritizing investments, while explicitly avoiding siloed arguments. It could capture the knowledge, needs, and voices of the full range of stakeholders, rather than just those with the most prominent or politically-connected platforms. Such a systematic approach might help rebuild the capacities and capabilities of, and trust in, America's public health institutions.

Edward W. Chan is a senior operations researcher and Andrew M. Parker is a senior behavioral and social scientist at the nonprofit, nonpartisan RAND Corporation.

Commentary gives RAND researchers a platform to convey insights based on their professional expertise and often on their peer-reviewed research and analysis.